The broad long-term objective of this research proposal is to provide a non x-ray method for detecting and characterizing breast cancer in the mammographicaIly dense breast. The achievement of this goal would alleviate the contrast problems radiologists currently face in reading x-ray mammograms of the young dense breast, and improve the diagnosis and characterization of breast cancer. The benefits of this research plan would include the reduction of mammographic false positives and a reduction in surgical biopsy rates. The specific aims of the GE Ultrasound Tagging of Light for Breast Cancer Imaging project will be to - * Provide breast tissue characterization using optical absorption spectroscopy by correlating spectral signatures with pathology. * Develop a single mechanism for lesion location in three dimensions and photon localization in turbid tissue with millimeter resolution for small (about a few cm.) tissue volumes. * Combine tissue characterization with the 3D lesion location mechanism into a spectroscopy-at-a-voxel imaging system for small tissue volumes capable of determining the malignancy of lesions and their depth within the breast in the specific case of lesions coded as suspicious by x-ray mammography. Perform feasibility studies for extending the system to whole breast imaging. * Study the effect of focussed ultrasound on phased array density waves. [IRPG collaboration.] The Research Plan. A spectral discriminator of benign and malignant breast processes based on metabolic rate will be evaluated by applying visible to near IR absorption spectroscopy in vitro using breast biopsies and in vivo using human breast cancer xenografts. The optical results will be compared with PET metabolic rate determinations of human cancer xenografts in nude mice. GE's new technique of 'tagging' scattered light in tissue with the frequency of an ultrasound (US) pulse will be evaluated as a mechanism for lesion location and photon localization in turbid media. The focal spot size of the US pulse is expected to be the fundamental system resolution. The 'tagged' light will be analyzed spectrally to determine tissue characterization using the spectral discriminator found in the spectroscopic studies.